The cloud-native orchestrator for the whole development lifecycle, with integrated lineage and observability.

dagster

What is the difference of passing inputs via input_values versus run_config in the job_execution? If there is historical differences from old dagster to new please provide, thanks.

`input_values` allows you to provide any python type, but it is basically hard coding the value. providing input as run config allows you to change it per run, but you have to use types that are mungeable as config (primitives and python builtins basically)

<@U017KUAENS0> So tell me if I am correct here. pandas dataframes are non mungeable via the accepted datatypes? So they would need to be passed as input_values? Alternatively you can pickle it, and then the config would contain a path or some reference to that data frame stored on disk?